Methods Group Meetings
Thursdays 14:00 pm, CBU west wing seminar room unless otherwise stated.
Meetings currently planned in 2009
27 Aug - Vassilis Tsiaras will present 'two case studies applying graphical visualisation to biomedical networks (1) EEG synchronization and (2) Gene Ontology'.
25 Jun - Christine Law on 'Fast Non-Cartesian Imaging and Sparse Sampling in MRI'
11 Jun - 'Sensor analysis in SPM' Introduction by Olaf Hauk followed by discussion, addressing questions of how to handle the gradiometer data: Because we have two at each measurement location, and they should be treated together, "classical" random field theory apparently cannot be applied. Combining the signals (e.g. taking the root-mean-square of two gradiometers at each location) is one solution, but the resulting data are not normally distributed. We will look at transformations which make data more nearly gaussian and the resulting scaling issues and how to interpret the results.
Meetings in the past
MEG Masterclass 2-4pm THEORY in the Seminar Room PRACTICAL in the Computer Users Room
30 Apr - MEG Masterclass Session 7. Interpretation of source solutions PRACTICAL
16 Apr - MEG Masterclass Session 6. Source solutions PRACTICAL
02 Apr - MEG Masterclass Session 5. Source reconstruction – assumptions THEORY
05 Mar - MEG Masterclass Session 4. ICA THEORY / PRACTICAL
26 Feb - MEG Masterclass Session 3. RMS then Contrast OR Contrast then RMS THEORY
19 Feb - MEG Masterclass Session 2. 3D sensor SPMs PRACTICAL
12 Feb - MEG Masterclass Session 1. Pre-processing PRACTICAL
05 Feb - Apoorva Bhandari will lead a journal club discussion on the paper 'Representational similarity analysis - connecting the branches of systems neuroscience' by Kriegeskorte, Mur and Bandettini which is available at http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2605405
15 Jan - Journal club on the paper 'Voodoo Correlations in Social Neuroscience' by Edward Vul, Christine Harris, Piotr Winkielman, & Harold Pashler
20 Nov - John Dunn - 'Models of recognition memory: A state-trace analysis' The present paper addresses the current debate between univariate and bivariate models of recognition memory. Univariate models propose that a recognition decision is based on evaluation of a single strength-of-evidence dimension while bivariate models propose that the decision is based on the separate combination of two dimensions, often labeled as recollection and familiarity. Both of these models fit existing experimental data well and a definitive test has not yet been forthcoming. I apply state-trace analysis to this question and evaluate the crucial point that if the bivariate model is correct, there should be evidence of bidimensionality in the data. I report the results of four experiments and a meta-analysis of 37 published experiments which yield little or no evidence of bidimensionality. I conclude that the univariate model provides a consistent and parsimonious account of recognition memory and discuss the general application of state-trace analysis to other similar issues in memory, judgment and decision making.
6 Nov - Various Local MEG Gurus - 'MEG Data Analysis Support Session' Your chance to ask questions about the issues that bug you around analysing data from MEG studies.
30 Oct - Sharon Wang - 'DSI-Studio demonstration'
16 Oct - Rik Henson - 'DCM for fMRI and MEG/EEG'
10 July - Journal club on paper by Y. Iturria-Medina et al. 'Characterizing brain anatomical connections using diffusion weighted MRI and graph theory', Neuroimage 36(3). For those with permissions it's on the wiki at http://imaging.mrc-cbu.cam.ac.uk/methods/DiffusionPapers under 'Tractography'. They use graph theory (but not 'graphical models') probabilistic structural maps and Diffusion Weighted Imaging to look at probabilistic anatomical connectivity maps.
1 July - Tips for EEG MEG MRI data collection: because you only collect the data once. - DanielWakeman
23 June - Discussion on the different methods that have been used to identify DTI fibre tracks and to measure DTI connectivity between regions - JohnGriffiths (See http://imaging.mrc-cbu.cam.ac.uk/methods/ApproachesToUsingFiberTracking)
14 February - Code Version Control: MatthewBrett
31 January - OS Scripting in Python: MatthewBrett
17 January - CC Chen (FIL, London): DCM for induced EEG/MEG responses - MeetingDcmMegChen
- 13 December - TBA
- 6 December - discussion on MEG analysis pipeline
- 22 November - discussion of methods group strategy
23 August - Spin Echo fMRI - ChristianSchwarzbauer
20 July (provisional) High-resolution FMRI, and current options at the CBU - ChristianSchwarzbauer
6 July: SPM clinic - MatthewBrett
- 14 June: No meeting, HBM conference
1 June (NOTE this is a Friday rather than a Thursday): Talk, Said Boujraf, Ultra fast BOLD fMRI Using Single Shot Spin-Echo EPI With SENSE at 3 Tesla. See SaidBoujrafAbstract.
17 May: Whence methods? - information gathering about current and future methods support and development - IanNimmoSmith
10 May: Discussion of diffusion analysis - MatthewBrett